Terry Hancock wrote:
James Richard Tyrer wrote:
Jon Smirl wrote:
> The fonts are drawn 3X as wide as you want. After filtering the 3x
> width is collapsed down into a single pixel which converts to to a
> three component alpha channel. The three components correspond to
> the RGB pixels of LCD sub-pixel rendering.
>
> There is a paper about this on the Microsoft site.
>
Since this would seem to be impossible, it would be a very
interesting paper to read.
Subpixel rendering, you mean?
Subpixel rendering works on *digital* display panels.
But what isn't going to work is rendering smaller screen fonts from a
large monochrome bit mapped font using a two dimensional FIR filter or
wavelet. Down sampling of pixmaps always results in a loss of sharpness
and a sharpness enhancement filter or wavelet isn't going to fix it
because down sampling results in a loss of information. Such a filter
does not restore missing information; it only gives the illusion of
doing so by increasing local contrast near edges. Hence the proper name
artificial sharpness enhancement.
Also, smaller fonts are hinted for optimum display at lower screen
resolutions. The absolute best that down sampled fonts can look is the
way anti-aliased WYSIWYG fonts look.
--
JRT
_______________________________________________
Open-graphics mailing list
[email protected]
http://lists.duskglow.com/mailman/listinfo/open-graphics
List service provided by Duskglow Consulting, LLC (www.duskglow.com)
